I'm pretty sure this hasn't really been done before, at least in TI-BASIC, but if something similar exists, please notify me.

Last year, I was absolutely amazed by the simplicity and ingenuity of Google Calculator, especially where you can type in any expression using any units (and an optional unit to convert to), and I decided that that was something someone needed to code for TI calculators. Thus I went on a coding rage, and actually got much of it done within a month, to the point where you could type in something like 7 M * IN INTO CM CM and it would display 1778. Unfortunately, I couldn't figure out division, parentheses, or exponents, so I set it aside and continued with other projects. One day, I really needed to clear some ARC space, so I deleted the group where I had saved it (I called it CALCALCA). I finally started regretting this decision, and will start coding it again, but I really need help with the issues I mentioned above.

This is what I've decided for the program so far:
1. It will use a blank input (Input "",Str1) to input a continuous expression.
2. The resultant string will searched through first for illegal characters.
3. The string is checked a second time, this time for the characters " INTO " (which specifies that the following units say what unit the user wants)
4. First, "(" is added to the beginning, and ")" to the end, just for padding (i.e., "("+Str1+")"→Str1)
5. The string is then evaluated, one character at a time.
    a. Whenever it hits a space, it tests to see whether * follows it.
        i. If there is one, the space is deleted by using sub('s for the two sides of the string.
        ii. If there is not, the space is replaced with )(, which assumes the space meant implied multiplication.
    b. Whenever it hits a letter, a subloop gets the entire combination of letters, which is saved to another string for checking.
        i. The second string is checked against a database of all the units to be supported (see note below).
        ii. If there is a match, that entire section of the main string is replaced with the numerical value of the unit, surrounded by "(" and ")", and the unit's base units are added to a list (see second note below). matrix.
    c. Whenever it hits a number, ".", "*", or "E", it it added directly.
    d. Whenever it hits a "(", the matrix is widened by one, and base units added from then on are added to the new column.
    e. Whenever it hits a ")", it stored the rightmost column into L₁ and checks the following character.
        i. If it is a "^-1", L₁ is inverted.
        ii. If it is a "²", L₁ is doubled.
        iii. If it is a "^", another subloop is used to get the entire number following it, and L₁ is multiplied accordingly.
        iv. L₁ is integrated into the next-rightmost column of the matrix.
6. When this is all done, the program will evaluate the string (with an exp() and (theoretically) appends the final unit.

Note:
The database consists of a huge (literally, huge, and might even be larger than the program itself) string containing each supported unit in the format:
"   2            Pa           -1    01    00    -2             1.00000000           ..."
    ^            ^            |                  |                  ^                ^
    |            |            +------------------+                  |                |
    |            |                     |                            |                |
Length of     Unit name.    These four numbers are the       How many of the      Continue
unit name.                  base units for the unit, in      unit fit into        onto next
                            the order {length, mass,         the base unit.       unit.
                            current, time}. For example,     Since pascals are
                                Pa=N/(m^2)                   the base for
                                =(kg*m/(s^2))/(m^2)          pressure, this
                                =kg/(s^2)/m                  would be a one
                            so is mass (the positive         (the zeroes are
                            one) divided by time squared     because it needs
                            (the negative two) divided by    to have exactly
                            length (the negative one).       ten characters).

Anyone willing to help me with this program? I currently have several programs in development simulataneously, and I can't finish this myself.
By the way, if anyone can't read the above code, I attached it as a .txt:

I'll try to help. I tried to make the BASIC program that gives the derivative using string. Although I didn't finish it, developing it helped me understand a lot about strings and doing math stuff with the string.
But I don't have my calc atm, and will be able to help you after around mid-June.

EDIT: Just thought of something: The newest version of WabbitEmu supports exporting variables, right?

Well, I never used that feature so, I don't know for sure
You can post whatever you have until the time you'll be able to upload and I'll try to pick it up from there

Thanks!

Anyway, I'm currently working on the data string. Here's an example for the length functions I've included so far:

:"1m  010000001.000000002in 0100000039.37007872ft 010000003.280839902yd 010000001.093613302mi 010000006.21371E-4...

The problem I'm already seeing now is that I have to include 8 characters for the unit combination every time I add another unit, even if it's exactly the same as for the previous few units (in this case, 01000000). If I could get rid of most of them, it would free up a lot of memory, but then it'd be somewhat hard to access the unit type of an individual unit. Can anyone think of better way to do this?

EDIT: Biblical cubits are now supported for the fun of it.

If you're still working on this over at UTI and such I'm sure Goose or Thornahawk could help over there. They seem to be very math savvy (10 point word of the day ).

For mass I would change g to kg, well that's just me since I'm a physics guy.

Oh ok, sorry I missed that from earlier. I'm very tired.

Isn't there also m³ for volume? (cube meter)

That's what we often used in physics classes over here besides liters

I'm also treating operations (all operations, not just exponents) separately.

Actually, that's the part that stumped me when I first made this program a year ago. I could get multiplication and implied multiplication working fine, but not division, exponents, or parentheses. Do you think I should just simplify things by including squares and cubes as units on their own?